Issue No. 10 December 2005 www.hkbiodiversity.net Feature Article Contents page Roost Censuses of Cave Dwelling Bats Feature Article : of Hong Kong Roost Censuses of Cave Dwelling 香港穴棲性蝙蝠調查 Bats of Hong Kong 1 Chung-tong Shek and Cynthia S.M. Chan An Introduction to Two Exotic Mammal Working Group Mangrove Species in Hong Kong: Sonneratia caseolaris and S. apetala 9 漁農自然護理署於過去兩年的夏季和冬季，在香港各主要的 四十多個洞穴內進行了兩次洞棲性蝙蝠調查，當中包括引水隧道、 Working Group Column : 荒廢的礦洞、海蝕洞和大型的排水溝等，共發現十一種蝙蝠，其中 霍氏鼠耳蝠數目稀少和分佈狹窄，被列為可考慮優先加強保育的洞 Butterfly Garden in the Shing 穴性蝙蝠物種。 Mun Country Park 13

Introduction Discovery of the Fifth Seagrass Species in Hong Kong – Bats spend more than half of their lives in day roosts and their Halophila minor 16 roosting behavior is species-specific. Some inhabit several types of roost while others specialize in only one type of roost. Among the 26 recorded species in Hong Kong (Shek and Chan, 2006), 14 bat species aggregate in caves such as water tunnels and abandoned mines as their day roosts (Fig. 1). Roost census is ideal to record Contribution to the Hong Kong Biodiversity such cave dwelling species, as it is possible to estimate colony Do you have any views, findings and sizes and composition of bats in each cave. observations to share with your colleagues on the Biodiversity Survey programme? At past, the distribution and status of most bat species were Please prepare your articles in MS Word poorly known in Hong Kong. Ades (1999) found that many species were only recorded in one to three localities. To better understand format and send as attachment files by the bats and their roosting sites in Hong Kong, a long term email to the Article Editor. monitoring program by AFCD was launched in 2002. The roost census is part of the baseline ecological survey for monitoring the Subscribing Hong Kong Biodiversity cave dwelling bats in Hong Kong. If you would like to have a copy, or if you know anyone (either within or outside AFCD) who is interested in receiving a copy of this newsletter, please send the name, organization, and email and postal addresses to the Article Editor.

Chief Editor: P.M. So ([email protected]) Article Editor: K.Y. Yang ([email protected])

© All rights reserved.

Fig 1. Roost of bats in cave, the Himalayan Roundleaf Bat (Hipposideros armiger 大蹄蝠) in a water tunnel in Sai Kung.  Methods

Direct roost censuses of bats were done in their day roosts, included all major water tunnels, abandoned mines, sea caves, drainage culverts and air raid shelters, in summer (June to August 2004) and winter (December 2004 to February 2005) (Fig. 2). The number of settled individuals, usually in clusters or aggregations, was counted or estimated by determining the cluster densities in selected areas and extrapolating these by the total area of the colony covered by the settled bats. For flying individuals, the number of individuals passing through a reference point was counted. In most caves, the data were further confirmed by the nightly emergence counts and harp trap surveys. Fig 2. Site surveyed in the winter and summer censuses in 2004-05. Species identification generally followed the keys by Shek (2004). The Rufous Horseshoe Bat (Rhinolophus rouxi sinicus 魯氏菊頭蝠, 華東亞種), however, has been upgraded to an independent species as Chinese Horseshoe Bat (R. sinicus 中華菊頭蝠) (Thomas, 1997). Owing to the unclear status of the Common Bent-winged Bat (Miniopterus schreibersii 長翼蝠) of Hong Kong, all larger bent- winged bat with forearm length exceeded 46 mm was considered as the Greater Bent-winged Bat (M. magnater 大長翼蝠) in this survey.

Based on the percentage of the species out of the grant total of individuals of all bat species (% of species) and the percentage of sites at which the species were recorded (% of sites), the status of species is classified as follows: % of species % of sites Very Common > 5 % > 30% Common > 5 % 10 – 30 % Uncommon 0.1 – 5 % > 10 % Rare < 0.1 % < 10 % To reduce disturbance to the bats, each visit was limited to two persons, and caves were visited no more than once per month, as bats subject to frequent disturbance may be forced to wake up from hibernation or torpor, perhaps leading to increased mortality by depletion of energy reserved, especially during winter when food supply is limited.

Results Species Comparison

At least 11 species of bat were recorded in this study. The total number of individuals and number of sites at which a species was recorded were shown in Table 1. The total number of bats recorded in the summer and winter censuses were 12,987 and 21,178 respectively. The Himalayan Roundleaf Bat (Hipposideros armiger 大蹄蝠) was the most abundant species in the summer census whereas the Leschenault’s Rousette (Rousettus leschenaulti 棕果蝠) was most abundant in the winter census. In both censuses, Chinese Horseshoe Bat was the most widely distributed species. The status of 11 species recorded in the censuses are ranked in Table 1, in which, the Horsfield’s Bat (Myotis horsfieldi 霍氏鼠耳蝠) was the only rare cave dwelling species in Hong Kong that was found in the water tunnels in Shek Kong and Nam Chung.

Table 1. Bat species recorded in the winter and summer censuses in 2004-05. Species Name Site * Estimated number** Status # Summer Winter Summer Winter Himalayan Roundleaf Bat 20 (66.7%) 19 (45.2%) 4332 (33.6%) 4392 (20.7%) Very Common Pomona Roundleaf Bat 11 (36.7%) 15 (35.7%) 1962 (15.2%) 3739 (17.7%) Very Common Chinese Horseshoe Bat 21 (70.0%) 25 (59.5%) 1744 (13.5%) 1416 (6.7%) Very Common Leschenault’s Rousette 5 (16.7%) 10 (23.8%) 870 (6.7%) 8828 (41.7%) Common Greater Bent-winged Bat 9 (29.0%) 24 (57.1%) 1879 (14.6%) 1439 (6.8%) Common Rickett’s Big-footed Bat 5 (16.7%) 7 (16.7%) 1527 (11.8%) 258 (1.2%) Common Lesser Bent-winged Bat 4 (13.3%) 13 (31.0%) 309 (2.4%) 457 (2.2%) Uncommon Intermediate Horseshoe Bat 11 (36.7%) 18 (42.9%) 144 (1.1%) 431 (2.0%) Uncommon Least Horseshoe Bat 8 (26.7%) 20 (47.6%) 39 (0.3%) 163 (0.8%) Uncommon Chinese Myotis 9 (30.0%) 15 (35.7%) 86 (0.7%) 47 (0.2%) Uncommon Horsfield’s Bat 3 (9.9%) 1 (4.8%) 5 (0.04%) 8 (0.04%) Rare * Total number (percentage) of sites at which the species was recorded. ** Estimated number of individuals of the species (percentage of the species out of the grant total of individuals of all bat species) recorded. # See text for details on status.  Site Comparison

A total of 42 sites were visited in the surveys. More sites were visited during the winter census because some caves were inaccessible during the rainy season, i.e. summer. The water tunnels SK01 (Sai Wan) and KTS01 (Kau To Shan) scored the highest number of bats in the summer and winter censuses respectively. The Himalayan Roundleaf Bat and Leschenault’s Rousette were the major species found in these two tunnels. Highest number of species were recorded in the water tunnel SK02 (Pak Sha O) with nine and eight species in the summer and winter censuses respectively.

Water tunnels situated in nearby areas was considered as a single tunnel system in this study. Data obtained from the water tunnels in Sai Kung (SK01-02), Kau To Shan (KTS01-02), Shek Kong (TLC01-04), Lantau North (LT01-02) and Plover Cove (PC01-03) were thus grouped in the site comparison (Table 2). Both Sai Kung and Kau To Shan tunnel systems scored high abundance and species richness of bats are ranked as the key roost caves (water tunnels) of bats in Hong Kong. For the four abandoned mines, the abundance and species richness were relatively lower than that of water tunnels, and the bats found were usually dominated by one or two species (Table 3). Over 1,000 individuals of Pomona Roundleaf Bat (Hipposideros pomona 小蹄蝠) were found in Lin Fa Shan while the Lin Ma Hang mine was the major roost for two bent-winged bats: the Lesser Bent-winged Bat (Miniopterus pusillus 南長翼蝠) and Greater Bent-winged Bat. Table 2. Major water tunnel systems surveyed in the study. Water Tunnel Site No. of Species in Abundance in Major Species Reference Summer / Winter Summer / Winter (Summer / Winter) * Sai Kung SK01-02 9 / 8 3486 / 2712 Himalayan Roundleaf Bat (56.9% / 75.8%), Rickett’s Big-footed Bat (26.4% / 4.9%), and Greater Bent-winged Bat (7.5% / 9.8%) Kau To Shan KTS01-02 9 / 9 1533 / 7039 Leschenault’s Rousette (34.2%/ 87.3%), Himalayan Roundleaf Bat (24.7% / 9.3%), and Pomona Roundleaf Bat (34.2% / 2.8%) Shek Kong TLC01-04 7 / 9 1429 / 1169 Chinese Horseshoe Bat (31.3% / 49.9%), Himalayan Roundleaf Bat (47.7% / 22.8%), and Greater Bent-winged Bat (5.2% / 12.6%) Lantau North LT01-02, 04 9 / 9 611 / 1320 Pomona Roundleaf Bat (39.8% / 32.3%), Greater Bent-winged Bat (20.9% / 10.2%), and Himalayan Roundleaf Bat (11.6% / 23.9%) Plover Cove PC01-03 5 / 9 928 / 1096 Himalayan Roundleaf Bat (80.6% / 3.2%), Pomona Roundleaf Bat (18.6% / 70.8%), and Greater Bent-winged Bat (0% / 7.4%) Nam Chung ** PC05-06 -- / 9 -- / 957 Himalayan Roundleaf Bat (-- / 52.5%), Pomona Roundleaf Bat (-- / 21.1%), and Greater Bent-winged Bat (-- / 7.8%) * Percentage of individuals of the species out of the total number of bat in the tunnel. ** The site was inaccessible during summer due to the strong currents.

Table 3. Abandoned mines surveyed in the study. Abandoned Site No. of Species in Abundance in Major Species Mines Reference Summer / Winter Summer / Winter (Summer / Winter) * Lin Ma Hang LMH01-02 3 / 6 1146 / 898 Greater Bent-winged Bat (81.7% / 73.3%), Lesser Bent-winged Bat (17.5% / 24.1%), and Chinese Myotis (0.8% / 0.7%) Lin Fa Shan LFS01-02 3 / 6 1396 / 1593 Pomona Roundleaf Bat (29.7% / 70.3%), Chinese Horseshoe Bat (21.5% / 70.3%), and Himalayan Roundleaf Bat (0% / 7.7%) Silvermine Bay SB 4 / 6 318 / 407 Himalayan Roundleaf Bat (51.9% / 57.5%), Pomona Roundleaf Bat (39.9% / 22.1%), and Chinese Horseshoe Bat (6.3% / 1.7%) Ma On Shan MOS01-02 1 / 3 2 / 72 Chinese Horseshoe Bat (0% / 79.2%), Intermediate Horseshoe Bat (100% / 0%), and Least Horseshoe Bat (0% / 18.1%) * Percentage of individuals of the species out of the total number of bat in the tunnel. Discussion

The roost censuses provide baseline information on the abundance and distribution of 11 species of cave dwelling bats in Hong Kong. Six bats species are ranked either very common or common in this study, including the ones which were thought to be either rare or uncommon in Hong Kong, such as the Rickett’s Big-footed Bat (Myotis ricketti 大足鼠耳蝠) and Greater Bent-winged Bat. Most roosts are either water tunnels or abandoned mines. These man-made structures provide excellent roosting habitats for many cave-dwelling bats in Hong Kong.

Among the species recorded, only the Horsfield’s Bat is ranked as “Rare”, worthy for further conservation enhancement. The water tunnels in Sai Kung (SK01 & SK02) and Kau To Shan (KTS01-KTS02) are the key sites of cave dwelling bats in Hong Kong.  Species Account

Himalayan Roundleaf Bat (大蹄蝠) – Very Common Hipposideros armiger (Hodgson, 1835) (Fig. 3) Colony size ranges from several individuals to a thousand or more, and individuals usually maintain a small distance between each other. The species can be found in various types of caves, such as abandoned mines, water tunnels and sea caves (Lamma Island). Large colonies were recorded in Silvermine Bay, Shek Kong, Nam Chung, Kau To Shan and Sai Kung (>1,300 individuals in January 2005) (Fig. 4).

Fig 3. Himalayan Roundleaf Bat (Hipposideros armiger 大蹄蝠)

Fig 4. Distribution of the Himalayan Roundleaf Bat in Hong Kong (2004-05).

Pomona Roundleaf Bat (小蹄蝠) – Very Common Hipposideros pomona Andersen, 1918 (Fig. 5) Colony size ranges from 50 individuals to more than 1,000, and individuals usually maintain a small distance between each other. The species roosts in various types of caves, and aggregates in small chambers or enclosures, like drainage pipes and air raid shelters. Large colonies were found in Shek Pik, Lau Shui Heung, Shan Liu, and Lin Fa Shan (>1,100 individuals in March 2005) (Fig. 6).

Fig 5. Pomona Roundleaf Bat (Hipposideros pomona 小蹄蝠)

Fig 6. Distribution of the Pomona Roundleaf Bat in Hong Kong (2004-05).  Chinese Horseshoe Bat (中華菊頭蝠) – Very Common Rhinolophus sinicus Andersen, 1905 (Fig. 7) This species forms tightly packed aggregation during winter, and sometimes, mixes with Intermediate Horseshoe Bat in their roosts. Colony size ranges from a few individuals to several hundreds. Strong sexual segregation was observed, at least during the breeding season. Females tend to form large aggregations during parturition and lactation and males usually roost alone or in small groups. Large colonies were found in Shek Kong, Lin Fa Shan and Sai Kung (over >1,000 individuals in May 2005) (Fig. 8). This species was once being considered as a subspecies of R. rouxi. However, based on the differences in the chromosome numbers (2n =36) of the R. rouxi sinicus, it is now upgraded as a separate species.

Fig 7. Chinese Horseshoe Bat (Rhinolophus sinicus 栗黃菊頭蝠) Fig 8. Distribution of the Chinese Horseshoe Bat in Hong Kong (2004-05).

Leschenaulti’s Rousette (棕果蝠) – Common Rousettus leschenaulti (Desmarest, 1820) (Fig. 9) In Hong Kong, this fruit bat always roosts in twilight zone or near the entrance of water tunnels, or even in sea caves. Colony size ranges from a few individuals to more than several thousands. Upon disturbance, the whole colony may take flight and swoop out of the cave. Large colonies were recorded in Shek Kong, Kat O and Kau To Shan (>6,000 individuals in October 2004) (Fig. 10).

Fig 9. Leschenaulti’s Rousette (Rousettus leschenaulti 棕果蝠)

Fig 10. Distribution of the Leschenaulti’s Rousette in Hong Kong (2004-05).  Rickett's Big-footed Bat (大足鼠耳蝠) – Common Myotis ricketti (Thomas, 1894) (Fig. 11) This Asian fishing bat tends to roost in abandoned mines and water tunnels. Colony size ranges from a few individuals to 1,000. In winter, it forms small groups in small rock crevices. In summer, this species forms large aggregations with other species, e.g. the Greater Bent-winged Bat and Chinese Myotis. Large colonies were observed in Tung Tze and Sai Kung in summer (>750 individuals in August 2004) (Fig. 12).

Fig 11. Rickett's Big-footed Bat (Myotis ricketti 大足鼠耳蝠)

Fig 12. Distribution of the Rickett's Big-footed Bat in Hong Kong (2004-05).

Greater Bent-winged Bat (大長翼蝠) – Common Miniopterus magnater Sanborn, 1931 (Fig. 13) This species forms tightly packed aggregations in winter with individuals stacking on each other. The largest over-wintering population of this species was located in an abandoned mine at Lin Ma Hang, which was designated as a Site of Special Scientific Interest in 1994. Colony size ranges from a few individuals to several hundreds. Aggregations sometimes mix with the Lesser Bent-winged Bat and Rickett’s Big-footed Bat. Some individuals could be found in rock crevices or weep hole of water tunnels during winter. Large colonies were found in Tung Tze (summer only) and Lin Ma Hang (>900 individuals in August 2004) (Fig. 14).

Fig 13. Greater Bent-winged Bat (Miniopterus magnater 大長翼蝠)

Fig 14. Distribution of the Greater Bent-winged Bat in Hong Kong (2004-05).

Lesser Bent-winged Bat (南長翼蝠) – Uncommon Miniopterus pusillus Dobson, 1876 (Fig. 15) Similar to Greater Bent-winged Bat but colony size ranges from a few individuals to several dozens only. It always forms aggregations with the Greater Bent-winged Bat during winter. Colonies were found in Tung Chung and Lin Ma Hang (>250 individuals in August 2004) (Fig. 16). 

Fig 15. Lesser Bent-winged Bat (Miniopterus pusillus 南長翼蝠)

Fig 16. Distribution of the Lesser Bent-winged Bat in Hong Kong (2004-05). Intermediate Horseshoe Bat (中菊頭蝠) – Uncommon Rhinolophus affinus Horsfield, 1823 (Fig. 17) Colony size ranges from few individuals to several dozens. This species preferentially hangs from ceilings or upper walls and maintains individual spacing, but individuals were also found roosting beside the colony of Chinese Horseshoe Bat. Sometimes, a few individuals may be found in the colonies of Chinese horseshoe Bat during winter. Colonies were recorded in Shek Kong, Fung Yuen and Shek Pik (91 individuals in February 2004) (Fig. 18).

Fig 17. Intermediate Horseshoe Bat (Rhinolophus affinus 中菊頭蝠)

Fig 18. Distribution of the Intermediate Horseshoe Bat in Hong Kong Least Horseshoe Bat (小菊頭蝠) – Uncommon (2004-05). Rhinolophus pusillus Temminck, 1834 (Fig. 19) Solitary individuals roost in scattered assemblages in caves. This species was also recorded in small chambers or drainage pipes. Usually, less than 10 individuals were recorded in each cave. Aggregations were recorded in Silvermine Bay, Tung Chung, and Shek Pik (46 individuals in February 2004) (Fig. 20).

Fig 19. Least Horseshoe Bat (Rhinolophus pusillus 小菊頭蝠)

Fig 20. Distribution of the Least Horseshoe Bat in Hong Kong (2004-05).  Chinese Myotis (中華鼠耳蝠) – Uncommon Myotis chinensis (Tomes, 1857) (Fig. 21) This species usually roosts alone or in pairs, but sometimes forms colonies of a few individuals to a dozen. In winter, it forms small groups in small rock crevices, but in summer, this species forms large colonies with other species, e.g. the Greater Bent-winged Bat and Rickett’s Big-footed Bat. Aggregations were recorded in Shek Kong, Lin Ma Hang and Sai Kung (35 individuals in August 2004) (Fig. 22).

Fig 21. Chinese Myotis (Myotis chinensis 中華鼠耳蝠)

Fig 22. Distribution of the Chinese Myotis in Hong Kong (2004-05).

Horsfield’s Bat 霍氏鼠耳蝠( ) – Rare (Species of Conservation Concern) Myotis horsfieldi (Temminck, 1840) (Fig. 23) It roosts in water tunnels, and spends the day secretly in weepholes alone or in small groups of 2 to 3 individuals. Records were found in Nam Chung and Shek Kong (6 individuals in December 2004) (Fig. 24). This Myotis species is so-called a sibling species of some other big-footed myotis, in which, species are physically similar to each other. The taxonomy of an individual captured was further confirmed by an mtDNA analysis (Lin and Chou, personal communication) and it is further confirmed by Dr. Gabor Csorba from the Hungarian Natural Museum.

Fig 23. Horsfield’s Bat (Myotis horsfieldi 霍氏鼠耳蝠)

Fig 24. Distribution of the Horsfield’s Bat in Hong Kong (2004-05). Acknowledgements Natural History Society 22: 183-209. Thanks are due to Prof. L. K. Lin and his student Shek, C.T. 2004. Bats of Hong Kong: An introduction of Hong Mr. C. H. Chou at the Tunghai University, Taiwan, Dr. Kong bats with an illustrative identification key. Hong Kong Armstrong Kyle at the Kyoto University Museum, Japan, Biodiversity 7: 1-9. and Dr. Gabor Csorba at the Hungarian Natural Museum Shek, C.T. and Chan, C.S.M. 2006. Three new bat species to Hong for their help and suggestions regarding the taxonomy Kong. Hong Kong Biodiversity (in press). and the molecular studies on bats. Thomas, N.M. 1997. A systematic review of selected Afro-Asiatic Rhinolophidae (Mammalia: Chiroptera): an evaluation of References taxonomic methodologies. Unpublished PhD. Thesis, Harrison Ades, G.W. J. 1999. The species composition, distribution, and Zoological Museum, Sevenoaks. population size of Hong Kong bats. Memoirs of the Hong Kong  An Introduction to Two Exotic Mangrove Species in Hong Kong: Sonneratia caseolaris and S. apetala

Winnie P. W. Kwok, Wing-Sze Tang and Barry L.H. Kwok Introduction Classification of Sonneratia

In early 2000, AFCD staff found some extra-ordinary Sonneratia are evergreen trees with open mangrove plants on the exposed mudflat close to the spreading crown which can grow up to 20 m high. mouth of Shenzhen River in the Deep Bay area. They Although they do not have buttress or prop roots, were later identified as an exotic mangrove species they have thick, cone-shaped and upright densely belonging to the genus Sonneratia. congregated pneumatophores which originate from the underground cable roots similar to the native Avicennia In 2001, Sonneratia individuals were also found marina. Flowers are solitary with numerous stamens among the native mangrove species including Aegiceras and vestigial (or no) petals. Fruits are fleshy and globule corniculatum, Kandelia obovata and Acanthus ilicifolius (Hogarth, 1999). Sonneratia spp. mainly distribute along the embankment of the downstream section of the in the tropical and subtropical area, from East Africa Kam Tin Main Drainage Channel (MDC) (Fig. 25). The through Indo-Malaya to tropical Australia, Micronesia 13 ha mangroves along the MDC were planted in 1998 and Melanesia including the Hainan Island of Mainland as a mitigation measure for the mangrove lost due to the China. MDC project. Being faster growing, Sonneratia could be easily recognized among other planted mangroves by The Sonneratia specimens found in the Deep Bay their height. As Sonneratia usually grow in the outermost area in 2000 were formerly misidentified as Sonneratia region of the seashore, their occurrence among native alba (杯萼海桑). Based on the key published by Hogarth mangrove suggests that they were planted mistakenly (1999), AFCD suggests that there are two species with the native species. In April 2002, the then Territory of Sonneratia in Hong Kong – Sonneratia caseolaris Development Department (now renamed as Civil (海桑) (L.) Engl and Sonneratia apetala (無瓣海桑) Engineering and Development Department) which Buch.-Ham (Table 4). In 2005, the Royal Botanic commissioned the original planting cleared Sonneratia Gardens of Kew in Britain confirmed the specimens sent found along the MDC. to them as Sonneratia caseolaris.

The two subcommittees of the Wetland Advisory Both S. caseolaris and S. apetala belong to the Committee discussed the occurrence of Sonneratia in Family Sonneratiaceae (海桑科) which consists of Hong Kong in May 2001. Although the potential impact of two genera: Sonneratia L. f. (海桑屬) and Duabanga this exotic species on the native mangrove communities (八寶樹屬) which accounts to 12 species (Wang and was still unknown, both subcommittees agreed to the Chen, 2002). The genus Sonneratia can be divided into precautionary measure to remove the exotic mangrove two Sections: Section Sonneratia with capitate stigmas from the Ramsar site. Since then, AFCD has conducted and Section Pseudosonneratia with peltate stigmas. Sonneratia removal exercises regularly. There are six species and three varieties in the genus.

Fig 25. Sonneratia in Kam Tin Main Drainage Channel. 10 Table 4. The morphological characteristics of S.caseolaris and S. apetala.

Species Sonneratia caseolaris (L.) Engl Sonneratia apetala Buch. -Ham

Chinese Name 海桑 無瓣海桑

Section Sonneratia Pseudosonneratia Height up to 15 m up to 20 m

Natural Distribution South East Asia to the northern Australia. Native to China South Asia such as India, Bangladesh and Malaysia. where they are naturally found on Hainan Island.

Leaves Broad, ovate, opposite leaf, incompletely unrolled to show Narrow, elliptical, opposite, gradually taper toward the venation. Apex acute in young plants which becomes round apex. Petiole is longer (~1 cm). at later stage. Length to width ratio is less than two. Petiole short (0.5 cm) and red.

Flowers Relatively larger (~5 cm). Bisexual. Stamens red and white Flower small (1.5-2 cm) and stalked, single or clustered distally, standing erect. Style greenish and long, twice the at branch ends. 4 green calyx lobes. Bisexual. Stamens length of stamens, topped with capitate stigma. Usually white, standing erect. Style yellow topped with mushroom- have 6 sepals. Petals red and oblong. Flowers appear all shaped or peltate stigma. Flowers appear from May to year round. December.

Fruits Compressed and edible fleshy. Large (up to 8.5 cm). Green Oval berry. Distinctively smaller (1.5-2.5 cm) than when young and turns yellowish green and aromatic when S. caseolaris. 4-5 calyx lobes. Green when young and mature. Produces 800-1,300 seeds per fruit. Fruits appear becomes grayish when mature. Each fruit produces all year round. 100-130 seeds. Fruits appear from August to early Spring. 11 Sonneratia in Afforestation Potential “Impact” of Sonneratia on Native Mangrove Because of their fast growing nature and the occurrence in the low intertidal level, Sonneratia are Field surveys carried out by AFCD in 2005 found often used to stabilize coastline or mudflat (Zan et al., that Sonneratia are distributed in the intertidal area 2003). In the ninety centuries, Sonneratia have been and channels of the Deep Bay area, Tsim Bei Tsui, widely adopted for afforestation in South China including Hong Kong Wetland Park, Nam Sang Wai, Sheung eastern and western Guangdong Province and Fujian Pak Nai and Ha Pak Nai (Fig. 27). There were about Province (Chen et al., 2003). 1,600 Sonneratia plants in the Deep Bay area (70% S. caseolaris and 30% of S. apetala). They are mainly In 1993, S. caseolaris and S. apetala were distributed in the outlets of stream and channels where introduced to the Futian Mangrove Nature Reserve salinity is relatively lower. They were also found in (福田紅樹林自然保護區) from Dongzhaigang Mangrove the outermost region of the mangrove forests or in Nature Reserve (東寨港紅樹林區) of Hainan Island isolated ‘gaps’ within the mangrove forests where light as part of a national key project to afforest Shenzhen penetrates. Bay. While S. caseolaris is native to Hainan Island, S. apetala was originally introduced from Sundarban in the southwest of Bangladesh to Hainan Island in 1985 (Liao et al., 2004; Zan et al., 2003).

In Futian, S . a p e t a l a a n d S. caseolaris started producing flowers and fruits successfully in 1996 and 1997 respectively (Zan et al., 2003; Li et al., 1998). Similar to those on Hainan Island, the tallest S. apetala reached 12.5 m. As at 2003, there was more than 0.6 ha of artificial forest of these two species in Futian.

Sonneratia are renowned for the ease to proliferate. The buoyancy nature of the seeds allows them to be drifted by the water current and travel a long Fig 27. Map showing the distribution of Sonneratia in the Ramsar Site. distance. Germination requirements are not specific and seeds tolerate a relatively wide There are some discussions on the potential impact fluctuation of temperature, pH and salinity (Liao of Sonneratia on the native mangrove flora community. et al., 1997). In view of these characteristics, the close When compared with other terrestrial and riparian proximity of the Deep Bay area in Hong Kong to the habitats, mangrove is relatively resistant to invasion as Futian Nature Reserve (Fig. 26), it is likely that the the instable substratum and saline conditions prevent Sonneratia in Hong Kong was originated from Futian. invasion of most exotic plant species (Teo et al., 2003). Zan et al. (2003) suggested that the niche of Sonneratia spp. does not overlap with the native mangroves and therefore it was unlikely that the indigenous mangrove species would be replaced. Moreover, Li et al. (2003) showed that Sonneratia spp. improve soil quality in the mudflat by increasing salinity, organic materials as well as nitrogen, phosphate and potassium concentration and reducing pH, which could promote the growth of native species. In fact, although Sonneratia spp. have colonized the Deep Bay area for several years, there is no sign that they have out grown the well established native mangrove species.

Fig 26. Map showing Mai Po Nature Reserve and Futian Nature Reserve. 12 Information Gaps and Way Forward Aegiceras corniculatum and Avicennia marina to determine their potential threat to the native Information on the basic biology and ecology and mangroves. the roles played by Sonneratia in Hong Kong is limited but at least it has been observed that some waterbirds n Evaluation of the different physical and chemical like grey herons or egrets made use of Sonneratia control methods to find out the most effective way to for resting. Sonneratia have been reported to serve clear these species. a number of ecological functions. In Singapore, the fragrant flowers of Sonneratia are pollinated mainly by Hopefully, more information would be available to the Dawn Bat (Eonycteris spelaea), the Common Long- help formulating an effective way to control these species tailed Bat (Macroglossus minimus), and the Lesser after the study. Meanwhile, the precautionary measures Short-nosed Fruit Bat (Cynopterus brachyotis). These to clear these exotic species would be carried out in the bats feed on nectar and pollen of flowers and rely mainly Ramsar site. on Sonneratia for sustenance (Tan, 2001). Similar to other native mangrove trees, many mangrove creatures References and plants also depend on Sonneratia. Being a pioneer Chen, Y.J., Liao, B.W., Peng, Y.Q., Xu, S.K., Zheng, S.F. and Zheng, D.Z. species that grow low on the tidal mudflats, Sonneratia 2003. Researches on the northern introduction of mangrove species stabilize riverbanks and coasts and provide a more Sonneratia apetala Buch-Ham. Guangdong Forestry Science and favourable ground for other types of trees and plants Technology 19(2): 9-12. (Chen et al., 2000). However, as Sonneratia are exotic Chen, Z.I., Wang, R.J. and Miao, Z.B. 2000. Introduction of SONNERATIA to Hong Kong and they grow faster than indigenous Species in Guangdong Province, China. In International Workshop mangrove species, there are concerns on their possible Asia-Pacific Cooperation on Research for Conservation of Mangroves impacts on the native mangrove species. 26-30.

The existing ‘control’ method is on-site mechanical Hogarth, P.J. 1999. The Biology of Mangroves. Oxford University Press. removal which may not be very cost effective because 228pp. Sonneratia spp. regenerate branches easily and quickly from cut trunks. Complete eradication could only be Huang, L. and Zhan, C.A. 2003. Analysis on introduction and trial of mangrove done by either removal of the young seedling when the Sonneratia apetala on the seashore of east Guangdong. Ecologic roots are not well established or dredging out the roots Science 22(1): 45-49. together with their extensive pneumatophores after cutting. In view of the vast area of mangrove and mudflat Li, M., Liao, B.W. and Zheng, S.F. 2003. Ecological effect of Sonneratia in the Deep Bay area and that complete root removal apetala plantation. Shanghai Environmental Sciences 22(8): 540-543. would inevitably affect the native mangrove trees nearby, Li, Y., Zheng, D.H., Liao, B.W., Zheng, S.F., Wang, Y.J. and Chen, Z.T. 1998. eradication of this exotic species seems impossible by Preliminary report on introduction of several superior mangroves. Forest using simple mechanical removal methods. Research 11(6): 652-655. In view of the above, AFCD initiated a study on Liao, B.W., Zheng, S.F., Chen, Y.J., Li, M. and Li, Y.D. 2004. Biological the basic biology and potential impact with a view to characteristics and ecological adaptability for non-indigenous mangrove preparing a management plan for these two species in species Sonneratia apetala. Chinese Journal of Ecology 23(1): 10-15. collaboration with Prof. Nora Tam Fung-Yee of the City

University of Hong Kong in summer 2005. The study is Liao, B.W., Zheng, D.Z., Zheng, S.F., Li, Y., Zheng, X.R. and Huang, Z.Q. expected to last for two years and covers the following 1997. The studies on seedling nursing techniques of Sonneratia topics: caseolaris and its seedling growth rhythm. Forest Research. 10(3): 296-302. n Coverage and growth rate to determine the spread of these species. Teo, D.H.L., Tan, H.T.W., Corlett, R.T. Wong, C.M. and Lum, S.K.Y. 2003. Continental rain forest fragments in Singapore resist invasion by exotic n Germination requirements through experiments in plants. Journal of Biogeography 30: 305-310. the greenhouse to quantify the habitat requirement of these species. Wang, R.J. and Chen, Z.Y. 2002. Systematics and biogeography study on the family Sonneratiaceae. Guihaia 22(3): 214-219. n Quantifying the flowering and fruiting season and evaluation of the seedling vigor in different fruiting Tan, R. "Mangrove Apple Sonneratia alba Perepat(Malay)". 2001. 20 Dec. 2005 seasons for formulation of the best season for

mechanical removal. Zan, Q.J., Wang, B.S., Wang, Y.J. and Li, M.G. 2003. Ecological assessment on the introduced Sonneratia caseolaris and S. apetala at the mangrove n Field competitiveness of seedlings with native forest of Shenzhen Bay, China. Acta Botanica Sinica 45(5): 544-551. mangrove species including Kandelia obovata, 13 Working Group Column Butterfly Garden in the Shing Mun Country Park

Eric Y.H. Wong, Vinci P.K. Li, Phoebe W.C. Sze and Alfred K.C. Wong Butterfly Working Group Introduction Planting Materials

The Butterfly Working Group commenced this Baseline surveys were conducted at the site and trial project, to establish an open butterfly garden in its adjacent areas to provide inventories of the existing August 2003. The garden forms part of the Working plants and butterflies in the area. Such information Group’s conservation action plan, which aims to carry out is important to avoid haphazard garden planning. For in situ conservation of butterflies, particularly the instance, the presence of a nearby larval foodplants may make it unnecessary to add the plant to the garden, and species of conservation concern identified in our the butterfly inventory together with knowledge of their territory-wide ecological survey. The butterfly garden larval foodplants and habitat preferences also form the project also aims to enhance the public awareness of basis of the plant selection. butterfly conservation by providing a place for the public to see and experience encounters with live butterflies. Apart from the plant species which have proven In contrast to traditional butterfly houses, in which ability to attract certain butterflies, plants used by the butterfly species of conservation concern physical environment and biological components are were also selected. For instance, India Birthwort controlled to keep butterflies for public enjoyment, ( Aristolochia tagala , 印 度 馬 兜 鈴 ) a n d I l l i g e r a open style butterfly gardens are usually constructed for (Illigera celebica, 寬藥青藤), which are the larval butterfly conservation as well as nature education. The foodplants of the Birdwings (Troides spp., 裳鳳蝶屬) principle of an open style butterfly garden is to support and White Dragontail (Lamproptera curius, 燕鳳蝶) wild populations of local butterflies by providing plant respectively, were planted at the garden. species having specific relations to their life cycles. Overseas examples of open style butterfly gardens In addition to larval foodplants, nectar sources can include the Smithsonian Butterfly Habitat Garden in the also entice butterflies. Various factors, namely flowering United States, and the Ecological Education Garden period, nectar production and flower shape, were of the Taiwan Endemic Species Research Institute in considered in choosing the appropriate plantings at the Taiwan. garden in order to ensure good nectar sources in terms of both quantity and quality throughout the year. The Site

Site selection determines the success of an open style butterfly garden, in terms of both butterfly conservation and public education. The selection was guided by various criteria, where a suitable site should (i) be located within the protected areas which facilitate the protection and management of the garden; (ii) comprise habitats for the growth of nectar plants / larval foodplants ˭̂́˸ʳ˅ and for the colonization of butterflies; (iii) have a good reserve of butterflies; and (iv) be easily accessible to the ˭̂́˸ʳˆ ˭̂́˸ʳ˄ public. The Shing Mun Country Park is without doubt a ˭̂́˸ʳˇ good candidate, given its rich records of butterflies and strong popularity among country park visitors. Among several potential sites in the Shing Mun Country Park, a terrace of about 2,560 m2 was selected for the trial project (Fig. 28). The public can easily reach the site with a 20-minute walk from public transport. The site, lying on a slight slope facing east, consists of four land patches comprising various microhabitats which fit the needs of different butterfly groups. The dense Fig 28. Location map of the Butterfly Garden natural woodland surrounding the site and the scrub A total of 6,700 butterfly larval foodplants and nectar plantings bounding the terraced zones form windbreaks, plants of 31 species have been planted at the garden which are required by many butterflies, and which also since 2003. Table 5 lists the butterfly larval foodplants protect the garden from disturbance. and nectar plants at the garden. 14 Planting Arrangement Visitor Facilities

While maintaining a mosaic structure of habitats To enhance the role as a place for nature education, through applying low levels of disturbance has been visitor facilities in the garden include a pavilion and two proven effective in promoting species diversity at butterfly picnic tables in Zones 1 and 2 respectively. The areas gardens (Kocher and Williams, 2000), proper planting of interest in the garden are linked by a well aligned arrangement at the garden links closely with the level of footpath, along which a total of 17 interpretative plates future maintenance required. have been installed to introduce butterfly ecology and Plantings at the butterfly garden were arranged conservation, and also the local butterflies which may be to form clusters of blooms of a species, and different seen in the garden (Fig. 31 and 32). clusters were planted close together to made maintenance easier, and the garden more attractive to butterflies both visually and fragrantly. Furthermore, the herbaceous plants were placed in front of the shrubs in accordance with the rule of thumb in gardening, that is, shorter in front, taller at the back (Fig. 29).

Fig 31. Pavilion at the Butterfly Garden, Zone 1.

Fig 29. Plantings at the Butterfly Garden, Zone 1. The plantings, together with the physical characteristics in different parts of the garden, create microhabitats for a variety of butterflies. Trees including a number of citrus (柑橘屬) were planted along the woodland edge of Zones 1 and 4 for butterflies of the families Papilionidae (Swallowtails and Birdwings, 鳳蝶科), Nymphalidae (Nymphs, 蛺蝶科) and Danaidae (Crows and Tigers, 斑蝶科). The shaded edges of these zones are welcomed by Satyridae (Browns, 眼蝶科) and Amathusiidae (Duffer and Faun, 環蝶科). A slope in Zone 4 is left untouched and remains as an open grassy area, which is attractive to Hesperiidae (Skippers, 弄蝶科). Two bamboo scaffolds were also erected in Fig 32. Interpretative plates at the Butterfly Garden, Zone 2. this zone for the climbers – India Birthwort and Illigera (Fig. 30). Being more exposed, Zone 3 is planted with Butterfly Uses at the Garden 灰蝶科 shrubs to attract Lycaenidae (Blues, ) and Pieridae Compared to the 31 butterfly species representing 粉蝶科 (Whites and Yellows, ). Zone 2 was a picnic eight families recorded in the baseline surveys, the area, within which the piece of open grassland allows number of species recorded in the garden since butterflies to soak up the sunshine during the day. September 2003 has increased more than twofold to 79 species of nine families. They include species of conservation concern such as Golden Birdwing (Troides aeacus, 金裳鳳蝶), Common Rose and White Dragontail. Rare species Common Spotted Flat (Celaenorrhinus hinus, 白觸星弄蝶) and the rare vagrant Comma (Polygonia c-aureum, 黃鈎蛺蝶) were also recorded.

Fig 30. A bamboo scaffold at the Butterfly Garden, Zone 4. 15 Larval foodplants in the garden have been used is also required, as weeds including Mile-a-minute Weed by various butterfly species. Common Rose was (Mikania micrantha, 薇甘菊) otherwise would overwhelm found laying eggs on the India Birthwort in Zone 4. the plantings, in particular the herbs. Winged Cassia (Cassia alata, 翅莢決明) in Zones 1 and 3 are highly effective in attracting the Whites and Unexpected management problems were Yellows including Lemon Emigrant (Catopsilia Pomona, recognized after the planting work commenced. Some 遷粉蝶) and Mottled Emigrant (C. pyranthe, 梨花遷 plantings in the garden were damaged by stray cattle, 粉蝶). Immature Swallowtails, namely Great Mormon or by feral monkeys. Some of the branches of young (Papilio memnon, 美鳳蝶), Common Mormon (P. polytes, trees were broken by feral monkeys, and the garden 玉帶鳳蝶) and Red Helen (P. helenus, 玉斑鳳蝶), have was particularly vulnerable to stray cattle that fed on the been found on the Chinese Lemon (Citrus limonia, 黎檬), plantings and stayed in the garden at night. To prevent Pummelo (C. maxima, 柚) and Mandarine (C. reticulate, stray cattle from damaging plants in the garden, fence 柑橘) in Zones 1 and 4. and uni-directional gates have been erected along the boundary of the garden. Nectar plants in the garden are also used by many butterflies. Distinguished nectar sources include Overall, an adaptive management approach has Lantana (Lantana camara, 馬纓丹), Blood-flower been adopted to plan the future maintenance work, ( Asclepias curassavica , 馬 利 筋 ) a n d P e n t a s which is to keep the diverse habitats at the site suitable (Pentas lanceolata, 五星花), where butterflies can for in-situ butterfly conservation and, at the same time, feed on their clustered flowers with less time and to meet the visitors’ expectation of a garden in bloom for energy spent. Billygoat-weed (Ageratum conyzoides, human appreciation as well as nature education. 藿香薊), Ivy Tree (Schefflera hetaphylla, 鵝掌柴) and Prickly Ash (Zanthoxylum avicennae, 勒欓花椒) were Ways Forward highly attractive to the Crows during their flowering Overseas experiences suggest that a successful period in the fall and winter. butterfly garden may necessitate at least three years of Maintenance of the Garden observation, trial and error planning and planting (New, 1997). The Butterfly Working Group will closely monitor Plantings in the garden include various flowering the butterfly uses in the garden and its conditions and shrubs, herbs and climbers providing nectar and carry out maintenance work where appropriate. larval food sources to butterflies. Field observations, however, revealed that keeping these plants in good Given the experience gained from this butterfly condition, particularly in the dry season, needs intensive garden project and that more than 98% of the local maintenance. In order to make the butterfly garden butterfly species that are represented in the protected a long-term and manageable project, fine-tuning areas, which cover about 40% of the total land area of maintenance work is essential. In fact, enrichment and Hong Kong, there is great potential for establishing more replacement plantings of more self-sustaining species such appropriate sites in our protected areas for in-situ with similar ecological functions have been carried out, butterfly conservation. and the effects are so far satisfactory. Frequent weeding

Table 5. List of butterfly larval foodplants and nectar plants at the Butterfly Garden. Nectar Plants and Larval Foodplants Asclepias curassavica, 馬利筋 Michelia x alba, 白蘭 Zanthoxylum avicennae, 勒欓花椒 Oxalis corniculata, 酢漿草 Mussaenda erosa, 楠藤 Duranta erecta, 假連翹 Cassia alata, 翅莢決明 Citrus limonia, 黎檬 Osmanthus fragans, 桂花 Maesa perlarius, 鯽魚膽 Citrus maxima, 柚 Perilla frutescens, 紫蘇 Crateva unilocularis, 樹頭菜 Citrus reticulata, 柑橘 Hedychium coronarium, 薑花 Viburnum odoratissimum, 珊瑚樹 Litsea glutinosa, 潺槁樹 Barleria cristata, 假杜鵑 Ligustrum sinense, 山指甲 Clerodendrum japonicum, 赬桐 Schefflera hetaphylla, 鵝掌柴 Pentas lanceolata, 五星花 Microcos paniculata, 破布葉 Nectar Plants Lantana camara, 馬纓丹 Spilanthes grandiflora, 大花金鈕扣 Ageratum conyzoides, 藿香薊 Lantana montevidensis, 小葉馬纓丹 Salvia splendens, 一串紅 Lagerstroemia indica, 紫薇 Gomphrena globosa, 千日紅 Melia azedarach, 楝 Bidens alba, 白花鬼針草 Raphiolepis indica, 石斑木 (車梅輪) Vitex negundo, 黃荊 Ixora chinensis, 龍船花 Murraya paniculata, 九里香 Larval Foodplants Aristolochia tagala, 印度馬兜鈴 Bambusa multiplex var. riviereorum, 觀音竹 Boehmeria nivea, 苧麻 Ovate Tylophora, 娃兒藤 Toddalia asiatica, 飛龍掌血 Illigera celebica, 寬藥青藤 16 Acknowledgements The Butterfly Working Group cordially thanks the Hong Kong Lepidopterists’ Society for their advice on the selection of site and plantings, and for the permission to use their photographs on the interpretative plates. We also wish to thank the Kadoorie Farm and Botanic Garden for providing the Blood-flower and the Field Investigation Unit of the Hong Kong Herbarium for the India Birthwort and Illigera. Assistance of the Shing Mun Country Park Management Office in managing the vegetation and constructing the facilities in the butterfly garden is also very much appreciated. References Kocher, S.D. and Williams, E.H. 2000. The diversity and abundance of North American butterflies vary with habitat disturbance and geography. Journal of Biogeography 27: 785-794. New, T.R. 1997. Butterfly Conservation. Oxford University Press, Melbourne.

Discovery of the Fifth Seagrass Species in Hong Kong – Halophila minor Barry L.H. Kwok, Chun-pong Lam and Joseph K.L. Yip Coastal Community Working Group and the Hong Kong Herbarium 本署職員最近在西貢土瓜坪發現香港第五個海草 品種 ─ 小喜鹽草。及後在蠔涌、斬竹灣及小灘亦發現此 品種。

During our territory-wide seagrass monitoring in 2002 and a re-visit on 30 March 2005 to To Kwa Peng, the first and second authors noticed a patch of seagrass with spoon-shaped l e a v e s g r o w i n g Fig 34. Leaf of Halophila minor o n t h e m u d f l a t s a d j a c e n t t o t h e m a n g r o v e ( F i g . 33). The seagrass looked like Halophila Fig 33. Halophila minor o v a l i s ( 喜 鹽 草 ) but with a different leaf-vein pattern. A subsequent examination of previously collected samples revealed that it was Halophila minor (小喜鹽草). Re-examinations of specimens previously deposited in the Hong Kong Herbarium also confirmed the presence ofH. minor. This species was previously unknown to Hong Kong (Hong Fig 35. Leaf of Halophila ovalis Kong Herbarium, 2004). The Halophila ovalis recorded in To Kwa Peng as Both H. minor and H. ovalis process elliptical reported in Table 2 of Issue No. 8 should therefore be spoon-shaped leaves with cross veins. Since they recognized as H. minor. After the discovery of H. minor are similar morphologically, they are quite difficult to in To Kwa Peng, the species was later recorded in three differentiate in the field. A possible way to identify them other localities namely Ho Chung, Tsam Chuk Wan and is by carefully examining their leaf-vein pattern. In Siu Tan. A taxonomic treatment of the Halophila species general, the leaf of H. minor possesses 3 - 10 pairs of in the Hong Kong will be published elsewhere in the near cross veins at an angle of 70 to 90o to the main vein future. (Fig. 34) while that of H. ovalis possesses more than 12 pairs of cross veins with a sharper angle of 45 to 60o References: (Fig. 35). The Hong Kong Herbarium. 2004. Check List of Hong Kong Plants 2004. Agriculture, Fisheries and Conservation Department, Hong Kong.